search

ronikobrosly / causal-curve

A python package with tools to perform causal inference using observational data when the treatment of interest is continuous.
MIT License
271 stars 18 forks source link

Distribution for Count Data #43

Open athammad opened 2 years ago

athammad commented 2 years ago

Hi, Currently, the library allows modelling the GPS only assuming a Normal, Lognormal or Gamma distributed Treatment. Is it possible to include the Poisson and Negative-binomial distribution for Count Data?

Thank you

ronikobrosly commented 2 years ago

Hi @athammad , it should be possible! The GPS is just a conditional prob density function of the exposure given a set of covariates, and it shouldn't matter if you arrive at that through a normal, lognormal, gamma, or any other type of model.

If you're up for it, it would be great to get your help with it. You could fork the repo, create a branch on your forked repo, and then open a pull request. I'll review it and work with you.